Journal of the American College of Cardiology
© 2008 by the American College of Cardiology Foundation
Published by Elsevier Inc.
EDITORIAL COMMENT
Defibrillation Threshold
Testing in Implantable
Cardioverter-Defibrillators
Might Less Be More Than Enough?*
Anne B. Curtis, MD, FACC
Tampa, Florida
Since implantable cardioverter-defibrillators (ICDs) were
first introduced in the early 1980s, defibrillation threshold
testing (DFT) at the time of implantation has been considered mandatory because the results have been used to
predict the likelihood that these devices would successfully
terminate sustained ventricular tachyarrhythmias when they
occurred clinically. From the initial days of epicardial patch
placements through the early generations of transvenous
devices with monophasic shocks, failure to defibrillate, even
with high-output shocks, was not uncommon. In addition
to intraoperative testing, the performance of DFT testing 1
or 2 days after implantation but before hospital discharge
became routine practice. The data acquired occasionally led
to the need for system revision. With the advent of biphasic
shocks in ICDs, mean defibrillation thresholds were lower,
and it became much less common for system revision to be
necessary. With the ability to reverse polarity, adjust the tilt
of the defibrillation waveform, or change the vector for
defibrillation, the need for the addition of a subcutaneous
patch or electrode array to the system is very rare in clinical
practice today.
See page 551
The established reliability of current ICD systems has led
most electrophysiologists to abandon the practice of routine
DFT testing before hospital discharge. Even annual DFT
testing, once a common clinical practice, is performed
infrequently today. In fact, DFT testing after implantation
is usually reserved now for specific situations, such as a
concern about lead status or a change in drug therapy (e.g.,
*Editorials published in the Journal of the American College of Cardiology reflect the
views of the authors and do not necessarily represent the views of JACC or the
American College of Cardiology.
From the Division of Cardiovascular Disease, University of South Florida, Tampa,
Florida. Dr. Curtis is a consultant for and has received honoraria and grant support
from Medtronic, Inc.; has received honoraria from Boston Scientific; and is a
consultant for and has received honoraria from St. Jude Medical.
Downloaded From: https://content.onlinejacc.org/ on 10/01/2016
Vol. 52, No. 7, 2008
ISSN 0735-1097/08/$34.00
doi:10.1016/j.jacc.2008.05.016
the addition of amiodarone) in the setting of a previously
marginal DFT.
Although practice patterns in the surveillance of patients
with ICDs have evolved remarkably, there has been little
change to date in the approach to the patient at the time of
implantation. Defibrillation is a probabilistic phenomenon
in which the higher the shock strength, the higher the
likelihood of successful defibrillation. The most common
approach in clinical practice to approximating the DFT is to
induce ventricular fibrillation at least twice and defibrillate
the patient with an energy setting at least 10 J less than
the maximum output of the ICD. Given the shape of the
defibrillation curve, this method should ensure an adequate safety margin in the overwhelming majority of
patients, although it does not technically determine the
actual DFT (1). The rare need for system revision and the
proven reliability of ICDs in terminating spontaneous,
sustained ventricular tachyarrhythmias have called into
question whether DFT testing at the time of implantation is
still required. Yet, physicians have been reluctant to abandon it, despite mounting evidence that it may no longer be
necessary (2– 6).
In this context we have new data acquired from patients
enrolled in the parent SCD-HeFT (Sudden Cardiac Death
in Heart Failure Trial) (7). As reported previously, SCDHeFT was a primary prevention ICD trial that established
the superiority of single-chamber ICDs compared with
amiodarone or placebo in improving survival in patients
with symptomatic heart failure and left ventricular ejection
fractions ⱕ35% (8). In a SCD-HeFT substudy by Blatt et
al. (7), reported in this issue of the Journal, baseline DFT
data were available for 717 patients implanted with ICDs.
The DFT protocol in the study was straightforward: ventricular fibrillation was induced and 20-J shocks were
delivered. If defibrillation was achieved, ventricular fibrillation was reinduced and the effects of a 10-J shock were then
determined. If 20 J failed the first time, 30 J was delivered
during the second induction. Regardless of the outcome of
DFT testing, all patients were programmed with the first
shock set at 10 J above the DFT (unless the DFT was 30 J,
in which case 30 J was programmed), with the maximum
output for subsequent shocks and reverse polarity for the last
2 shocks. Defibrillation was achieved in all patients with the
use of shock strengths ⱕ30 J. In addition, 97.8% of the
patients had a DFT ⱕ20 J. The main findings of the study
were that there was no difference in survival between
patients who had a DFT ⱕ10 J and those who had a DFT
⬎10 J. First-shock efficacy for spontaneous events was
83.0%, with no significant difference according to baseline
DFT. Of the 31 patients who had failed first shocks in
follow-up, all survived the event, with 3 of the patients
subsequently dying in the hospital from progressive heart
failure. In most of these patients who survived a failed first
shock, the arrhythmia terminated spontaneously or subsequent shocks successfully terminated the arrhythmia.
558
Curtis
Editorial Comment
Of the 16 patients with a DFT between 21 and 30 J, only
3 patients had an appropriate shock in follow-up, all of
which were successful. Granted, this is a small number of
patients and events. However, there were over 700 patients
in this substudy, with a median follow-up of 45.5 months.
The 2.2% risk of an elevated DFT is comparable to what has
been seen in other studies. With that low level of risk for a
high DFT, and a markedly lower risk of an adverse outcome
in follow-up, a much larger study with a prolonged
follow-up would need to be done to detect a difference in
outcome, if indeed one exists. Whether such a slight
difference in outcome would be clinically meaningful is also
questionable.
It should be noted that defibrillation using shock energies
up to 30 J was successful in all patients in this substudy. It
is thus unknown what kind of outcome might be expected
in patients who failed defibrillation with the maximum
output of an ICD at the time of implantation. However, it
is clear that in patients similar to those tested in SCDHeFT, that finding would be exceedingly rare.
It is important to stress that the results of this study can
be applied only to similar patients who are receiving
single-chamber ICDs for primary prevention and are receiving standard-of-care background medical therapy for heart
failure. The results should not be extrapolated to patients
receiving cardiac resynchronization therapy, who often have
more advanced disease. Whether the results could be
applied to patients receiving dual-chamber ICDs is debatable because the reason for the atrial lead may be a history
of atrial arrhythmias or a rare need for pacing that may not
have an impact on defibrillation. The results should also not
be applied to patients who are receiving ICDs for secondary
prevention, in whom life-threatening ventricular arrhythmias have already occurred.
A recent decision analysis and Markov model found that
5-year survival was nearly identical with or without DFT
testing (9). Only if the annual risk of a lethal arrhythmia was
at least 5% was there a slight advantage to DFT testing, but
the incremental benefit was minimal. Another factor that
must be considered in recommendations regarding DFT
testing is the risk associated with this practice. Birnie et al.
(10) recently reviewed the experience in Canada over a
6-year period and found that among 19,067 patients with
ICD implants, 3 deaths, 5 strokes, and 27 episodes of
prolonged resuscitation, some of which were associated with
serious clinical sequelae, were directly attributable to DFT
testing.
The current study is not the first to challenge the
conventional wisdom of routine DFT testing (2– 6). Viskin
and Rosso (5) recently enumerated elegantly a myriad of
reasons for avoiding DFT testing, including the lack of
correlation between induced and spontaneous ventricular
fibrillation; the fact that many patients will never have
spontaneous ventricular fibrillation in follow-up; the reality
that successful DFT testing at implant is so likely that little
Downloaded From: https://content.onlinejacc.org/ on 10/01/2016
JACC Vol. 52, No. 7, 2008
August 12, 2008:557–8
information is gained as a result of the procedure in most
patients; the observation that spontaneous arrhythmias are
often ventricular tachycardia, which is easier to cardiovert;
and the fact that DFT testing is not without risk, among
others.
These observations coupled with the results of the present
study indicate that routine DFT testing in stable patients
receiving nonresynchronization ICDs for primary prevention is of little clinical value. Driven by these results, it
would be reasonable to consider whether practice guidelines
should be revised to reflect these findings, with 2 important
caveats. First, a mounting consensus that routine DFT
testing may be safely abandoned in selected patients does
not mean that it should be discontinued in all patients. We
do not have information on patients receiving cardiac
resynchronization devices and other specific situations as
mentioned previously to make recommendations at this
time. However, proof-of-concept has now been established
that routine DFT testing is unlikely to provide sufficient
information that impacts ICD programming in the overwhelming majority of patients, and it carries some slight
risk. Collectively, these data justify a prospective trial
randomizing patients receiving ICDs for any indication to
DFT testing versus no DFT testing, because less testing just
might be more than enough.
Reprint requests and correspondence: Dr. Anne B. Curtis,
University of South Florida, 12901 Bruce B. Downs Boulevard,
MDC 87, Tampa, Florida 33612. E-mail: acurtis@health.usf.edu.
REFERENCES
1. Barold SS, Herweg B, Curtis AB. The defibrillation safety margin of
patients receiving ICDs: a matter of definition. Pacing Clin Electrophysiol 2005;28:881–2.
2. Strickberger SA, Klein GJ. Is defibrillation testing required for
defibrillator implantation? J Am Coll Cardiol 2004;44:88 –91.
3. Russo AM, Sauer W, Gerstenfeld EP, et al. Defibrillation threshold
testing: is it really necessary at the time of implantable cardioverterdefibrillator insertion? Heart Rhythm 2005;2:456 – 61.
4. Swerdlow CD, Russo AM, Degroot PJ. The dilemma of ICD implant
testing. Pacing Clin Electrophysiol 2007;30:675–700.
5. Viskin S, Rosso R. The top 10 reasons to avoid defibrillation threshold
testing during ICD implantation. Heart Rhythm 2008;5:391–3.
6. Markowitz SM. To test or not to test during defibrillator implantation:
A reassessment of the conventional wisdom. J Cardiovasc Electrophysiol 2008;19:406 – 8.
7. Blatt JA, Poole JE, Johnson GW, et al., for the SCD-HeFT
Investigators. No benefit from defibrillation threshold testing in the
SCD-HeFT (Sudden Cardiac Death in Heart Failure Trial). J Am
Coll Cardiol 2008;52:551– 6.
8. Bardy GH, Lee KL, Mark DB, et al. Amiodarone or an implantable
cardioverter-defibrillator for congestive heart failure. N Engl J Med
2005;352:225–37.
9. Gula LJ, Massel D, Krahn AD, Yee R, Skanes AC, Klein GJ. Is
defibrillation testing still necessary? A decision analysis and Markov
model. J Cardiovasc Electrophysiol 2008;19:400 –5.
10. Birnie D, Tung S, Simpson D, et al. Complications associated with
defibrillation threshold testing: the Canadian experience. Heart
Rhythm 2008;5:387–90.
Key Words: sudden cardiac death y implantable
cardioverter-defibrillator y defibrillation y ventricular fibrillation.